FUNCTIONAL CHROMOPHORES FOR LIGHT-HARVESTING APPLICATIONS

(1)

UNIVERSITÀ DEGLI STUDI DI TRIESTE

XXVIII Ciclo del

Dottorato di Ricerca in Scienze e Tecnologie Chimiche e Farmaceutiche

PhD School in Chemical and Pharmaceutical Sciences and Technologies

FUNCTIONAL CHROMOPHORES FOR

LIGHT-HARVESTING APPLICATIONS

PhD Student

Tanja Miletić

PhD School Director

Prof. Mauro Stener

PhD Supervisor

Prof. Davide Bonifazi

(2)

(3)

(4)

(5)

I

Acknowledgments

…and also this moment arrived. Here I am, trying to write down all the acknowledgments for the wonderful people I met during this experience, those who contributed directly to this manuscript and those who were close to me during this period, hoping I will not forget nobody.

Firstly, I would like to specially thank my supervisor Prof. Davide Bonifazi for the opportunity he gave me to start this PhD and to work on different challenging and very interesting projects, for his guidance during these years and for his constant motivation and tremendous enthusiasm.

Many thanks go to Prof. Maurizio Prato and Prof. Tatiana Da Ros for welcoming me in the big Prato group. I will be always grateful for the opportunity I had to work in such international environment.

I would like to sincerely thank all collaborators that were involved in the FIRB project and made this thesis possible. In particular, many thanks go to Dr. Eleonora Pavoni, Prof. Nicola

Armaroli, Dr. Mirko Panighel, Dr. Giovanni di Santo, Prof. Paola Ceroni, Dr. Giacomo Bergamini and Prof. Giuseppe Brancato for the collaborative work and help. Moreover, I

would also like to thank for their work the people from the University of Salento, Lecce, in particular Dr. Andrea Listorti and Dr. Silvia Colella.

Special thanks go to Dr. Federica de Leo, Dr. Andrea Fermi, Prof. Stelios Couris and

Giannis Orfanos, for the large amount of work performed for the perylene project even if

part of it is not included in this thesis.

The work on carbon nanotubes would not have been realized without the inestimable help of Dr. Riccardo Marega and Dr. Caroline A. Ahad Hadad. Thank you for being always open to help. Many special thanks are going to Dr. Nicola Demitri not only for the X-ray measurements, but also for his great dedication, help and collaboration.

Particular thanks go to Maria Mercedes Lorenzo Gracia, Lou Rocard and Dr. Antoine Stopin for the proofreading of this manuscript and the tremendous help. Moreover, I have to deeply acknowledge Nicolas Biot for the precious and inestimable help with the assembling of this thesis. Thanks to all of you for the enormous support. I am not sure I would manage to do it without you.

(9)

V

I would like to thank all the people from Bonifazi Group, Maria, Francesco, Davide, Andrea

S., Andrea F., Cataldo, Lorenzo, Lou, Nicolas, Alex, Antoine and Rodolfo. Thank you for the

nice welcome in Cardiff and for creating a very enjoyable working environment. Many thanks go also to the members of the old COMS group. In particular, I would like to thank

Silvia, John and Dario for the short collaborative experience we had the opportunity to share.

The biggest thanks go to all the people from Prato’s Lab, past and present members, for enriching me in these years in so many different ways. In particular, thanks go to Michela

P. (thank you for managing us in so efficient way), Caroline, Jose, Marco, Dani, Nuria, Cristina, Maribel, Lorenzo, Silvia, Alexa, Federico, Alejandro (thanks to you, Alex, I will

never forget where the Erlenmeyer’s are), Andrea, Mimmo, Adrian, Anirban, Francesco,

Angela, Jeni (I simply adore you), Davide (O Da! Che te devo dire, grazie!) and all others

from C11 and Pharmacy. Thanks to all for the good time we shared and for the great people you are.

I would like to express my thanks to Jacopo. Your scientific enthusiasm, hard motivation together with your well-mannered behavior made me very glad to have the opportunity to work with you, even if for a short period. Undoubtedly, special thanks go to some people who shared with me this experience, who are not only great colleagues but adorable friends.

Arturo and Manuel, you are the best lab mates ever. The time I had in and out the lab with

you was simply great; thank you for the nice moments, for the unforgettable support, for your craziness, but above all, thank you for the sincere friendship.

This experience would not be the same without the chicas group. The relation between us is difficult to describe. Ana, Maria, Vale and Agnieszka thank you simply for being present and for the enormous support, nice moments, discussions, arguments but especially for believing in me when I was not. The folly of our group relies in the peculiar beauty of each of you and for this I particularly thank you. Agnieszka thank you twice, you know why! You are my master of life. Maria, thank you for always helping and supporting me, this thesis would not be possible without you. Muchas gracias de corazon! Very special thanks go to Giuly, for the precious friendship and for having been the most unique person I ever met.

(10)

VI

durante questo cammino e per avermi sempre creduto ed affiancato in qualsiasi scelta intrapresa.

(11)

VII

List of abbreviations

Å Angstrom

Abs Absorption

AFM Atomic force microscopy

AlCl3 Aluminium chloride

aq. Aqueous

B2pin2 Bis(pinacolato)diboron

BChl bacteriochlorophylls

BLA Bond length alternation

BINOL 1,1’-binaphthyl-2-2’-diol ºC Degree centigrade (0 ºC = 273.16 K) calc. Calculated CB Chlorobenzene Chl Chlorophylls CH2Br2 Dibromomethane CH2Cl2 Dichloromethane CH3CN Acetonitrile CH3SO3H Methanesulfonic acid Cm Centimeter CMD Concerted-metalation-deprotonation CNTs Carbon nanotubes

CuO Copper(II) oxide

Cu(OAc)2 Copper(II) acetate

d% Doping ratio

DDQ 2,3-dichloro-5,6-dicyano-1,4-benzoquinone

DIPEA N,N-Diisopropylethylamine

(12)

VIII

DMSO Dimethylsulfoxide

DNF Dinaphthofuran

DOS Density of states

Dtbpy 4,4-di-tert-butyl bipyridine

Eg Energy bandgap

EInt Energy due to intermolecular interaction

eq. Equivalent

ERes Aromatic stabilization resonance energy

ESI Electrospray ionisation

Esub Enery due to the push-pull effect

ET Electrontransfer

EtOAc Ethyl Acetate

eV Electronvolt (1eV = 1.602 x 10-19_J)

Fs Femtoseconds

FG Functional group

FRET Förster-type energy-transfer

FTO Fluorine-doped thin oxide

GNRs Graphene nanoribbons

H Hour

H2SO4 Sulfuric acid

HDI Hexarylenebis(dicarboximides)

HLG HOMO-LUMO gap

HOMO Highest occupied molecular orbital

HR High resolution

HTM Hole Transport Material

Hz Hertz (s-1₎

IR Infrared

iPrOH Isopropanol

(13)

IX

KOH Potassium hydroxide

LiTFSI Lithiumbis(trifluoromethanesulfonyl)imide

LUMO Lowest unoccupied molecular orbital

M Molar

MALDI Matrix-assisted laser desorption/ionisation

MeOH Methanol

MHz Megaherzt

Min Minute

MS Mass spectrometry

MWCNTs Multi-Walled Carbon nanotubes

Mw Microwave irradiation

N Number of units

NMI Naphtalenemonoimides

NIR Near infrared

Nm Nanometer

NMR Nuclear magnetic resonance

ODCB Ortho-dichlorobenzene

OLED Organic-light emitting diode

OFET Organic field-effect transistor

OSC Organic semiconductor

OTFT Organic thin-film transistor

PAH Polycyclic aromatic hydrocarbon

PBIs Perylene bisimides

Ps Picosecond

PDI Perylene diimide

Pd(OAc)2 Palladium(II) acetate

PC71BM phenyl-C71-butyric acid methyl ester

PhI(OAc)2 (Diacetoxyiodo)benzene

(14)

X

PMI Perylenemonoimide

PLM Photoluminescence mapping

PCE Power conversion efficiency

PPN Poly(peri-naphthalene)

PPV Poly(p-phenylenevinylene)

ppm Parts per million

PSCs Perovskite solar cells

P3OT poly(3-octylthiophene)

p-TsOH p-Toluenesulfonic acid

PXX peri-xanthenoxanthene

RC Reaction center

r.t. Room temperature

Sc(OTf)3 Scandium(III) triflate

SEM Scanning electron microscopy

SDBS sodium dodecyl benzene sulfate

Spiro-OMeTAD

Tetrakis(N,N-di-p-methoxyphenylamine)-9,9′-spirobifluorene

SWCNTs Single walled carbon nanotubes

TAT Tetra-aza-terrylene

t_Bu _Tert-Butyl

TCE Tetrachloroethane

TD terrylenediimide

Tf2O Trifluoromethanesulfonic anhydride

TFA Trifluoroacetic acid

TGA Thermogravimetric analysis

THF Tetrahydrofurane

t.l.c. Thin layer chromatography

UV-Vis Ultraviolet-visible

vHS van Hove singularities

(15)

XI

VT Variable temperature

ZnTPP Zn (II) tetraphenyl porpyrin

λ Ultraviolet-visible

ɛ Molar extinction coefficient

Φfl Fluorescence quantum yield

θ Dihedral angle

θp Pyramidalization angles

(16)

XII

Abstract

The photosynthetic system is regarded as the most sophisticated nanobiological machine in nature. It deploys a large numbers of pigments in antenna complexes to harvest light and funnel the resulting excited-state energy to the reaction centers. The choice of the light absorbing components is one of the crucial steps in the design of synthetic light-harvesting systems. In this respect, the preparation of new dyes with tunable opto-electronic properties remains a very challenging task for the development of highly efficient light-harvesting materials.

Before addressing the detailed investigations of this thesis work, in Chapter I, a briefly introduction on natural and synthetic antennas along with an outlook on the possible types of approaches towards the engineering of the HOMO-LUMO gap for the development of new dyes with unique opto-electronic properties is given to the reader. Moreover, a detailed outline of the manuscript is described.

This dissertation focuses on the design, synthesis and characterization, as well as investigations of the optical properties, of different series of new functional chromophores. Specifically, Chapter II addresses the design and synthesis of a series of well-defined oligo (p-phenylenevinylenes) bearing styryl units of different π-extension, namely molecules 2-1 and 2-2. A double Horner-Wadsworth-Emmons-type reaction has been exploited for the synthesis of the most π-extended molecule 2-1. Subsequently, newly prepared chromophores have been used as donor moieties for the preparation of donor-acceptor systems

(f-SWCNT-1 and f-SWCNT-2), in which single walled carbon nanotubes (SWCNTs) plays the role of

the acceptor unit. The choice of the latter is motivated by the high conductivity, strong mechanical, thermal and environmental resistance that characterize this carbon material. A covalent approach

via amidation reaction has

(17)

XIII

between the individual entities and evaluate their applicability as photoactive material in energy conversion systems.

On the other hand, Chapter III describes the synthesis of novel light absorbing molecules based on oxygen-containing π-extended polycyclic aromatic hydrocarbons (PAH), in which two polyaromatic hydrocarbon substructures are bridged through one or two O atoms. The main building block used for this purpose is the perylene molecule. High-yielding ring-closure key steps were exploited for the formation of either furanyl or pyranyl frameworks, depending on the reaction conditions, through intramolecular C-O bond formation. Based on the shape of the aromatic systems and the conjugation of the π fragments, the absorption and emission maxima of newly prepared O-doped π-extended PAH were finely tuned throughout the entire visible region.

(18)

XIV

Riassunto

Il sistema fotosintetico è considerato il sistema “nano”- biologico più sofisticato presente in natura. Utilizza un gran numero di pigmenti, disposti in sistemi antenna, che raccolgono la luce e la trasferiscono a centri di reazione specializzati. La scelta delle componenti che assorbono la luce è uno dei passaggi cruciali nella progettazione di sistemi antenna artificiali. A tale proposito, la sintesi di nuovi cromofori con proprietà optoelettroniche modulabili rimane una delle sfide maggiori per sviluppare materiali che raccolgono la luce con elevata efficienza.

Prima di discutere nel dettaglio la ricerca svolta in questo lavoro di tesi, nel capitolo I, al lettore verrà presentata una breve introduzione riguardante i sistemi antenna naturali e artificiali e i possibili tipi di approccio utilizzati per il design di nuovi cromofori dalle uniche proprietà optoelettroniche; in particolare la modifica della differenza di energia fra gli orbitali HOMO e LUMO, detta anche il bandgap. Inoltre, questo capitolo riporta anche i punti essenziali dell’intero manoscritto.

Questo lavoro di tesi si concentrerà sulla progettazione, sintesi e caratterizzazione nonché sullo studio foto-fisico di diverse classi di nuovi cromofori funzionali. Più precisamente, il capitolo II descrive il design e la sintesi di una serie di composti oligo(p-fenilene vinilene) ben definiti, presentanti componenti stiriliche di lunghezze diverse, cioè le molecole 2-1 e

2-2. A tale proposito, la

sintesi della molecola 2-1 è stata realizzata utilizzando la

reazione di

Horner-Wadsworth-Emmons.

Successivamente, tali

cromofori sono stati utilizzati come specie donatore per la

preparazione di sistemi

donatore-accettore

(19)

foto-XV

fisiche dei sistemi così preparati sono state studiate allo scopo di evidenziare i processi di

trasferimento di energia per risonanza tra le singole componenti e valutarne l’applicabilità come materiale foto-attivo in sistemi di conversione di energia.

Il capitolo III descrive la sintesi di cromofori basati su idrocarburi policiclici aromatici (IPA) dalla superficie aromatica estesa e contenenti atomi di ossigeno, in cui due sottostrutture IPA sono connesse tra loro tramite anelli contenenti uno o due atomi di ossigeno. L’unità principale usata per la sintesi di tali nuovi cromofori è il perilene. Passaggi chiave di ciclizzazione ad elevata resa, basati su condizioni di reazione diverse, sono stati utilizzati al fine di preparare anelli furanici e piranici, tramite formazione di legami C-O intramolecolari. In base alla forma del sistema aromatico e la coniugazione dei vari frammenti contenenti atomi di ossigeno, i valori massimi di assorbimento e di emissione dei nuovi cromofori sono stati finemente regolati attraverso tutto lo spettro del visibile.

(20)

(21)

1

1. General introduction

Energy is the most important issue of the 21st_century.[1]_{The world’s current energy}

requirements are far from being satisfied, and the solar energy represents the most abundant renewable energy resource available to us. There is a huge gap between the present use of solar energy and its enormous potential.[2]_{This potential is demonstrated by the high}

efficiency of natural photosynthesis, which provides the necessary chemical energy for almost all life on Earth and this since more than two billion years.[3]

Photosynthesis literally means “synthesis with light” and it is the process used by plants, algae and certain bacteria to harness energy from sunlight into chemical energy.[4]_The

photosynthetic process begins with the absorption of sunlight by specialized pigment-protein complexes (known as light harvesting complexes), that function as antennae for incident energy. This is followed by the funneling of the excitation energy within the antenna assembly to special sites called the reaction centers (RCs), in which the captured energy is converted into chemical energy by means of electron-transfer reactions (Figure 1.1.).

.

Figure 1.1. Schematic representation of antennas and reaction centers in photosynthetic systems. Antenna pigments collect the light and energy transfer processes deliver the excited state to the reaction center where electron transfer reactions store the energy.[5]

(22)

2

distinct optical and redox properties) that enables maximum light absorption and energy conversion.[3,6]

The photoinduced energy transfer processes in most photosynthetic antennas are occurring by Förster resonance energy transfer (FRET) mechanism.

D + hv  D* D* + A  D + A*

A*  A + hv’

FRET is a distance-dependent interaction between the electronic excited states of two pigment molecules in which the excitation is transferred from a donor (D) molecule to an acceptor (A) molecule in a non-radiative fashion through long-range dipole-dipole interactions. In the process of FRET, initially a donor (D) fluorophore is excited by a photon and then relaxes to the lowest excited state, S1. The energy released when the electron returns

to the ground state (S0) may simultaneously be transferred to a nearby chromophore, the

acceptor (A). After excitation, the excited acceptor emits a photon and return to the ground state, if another quenching state do not exist. (Figure 1.2) The theory supporting energy transfer is based on the concept of treating an excited fluorophore as an oscillating dipole that can undergo an energy exchange with a second dipole having a similar resonance frequency.

Few criteria must be satisfied in order for FRET to occur: (i) the fluorescence emission spectrum of the donor molecule must overlap the absorption or excitation spectrum of the acceptor chromophore; (ii) the two chromophores (donor and acceptor) must be in the close proximity one to the other (typically 1 to 10 nanometer); (iii) the transition dipole orientation of the donor and acceptor must be approximately parallel to each other; (iv) the fluorescence lifetime of the donor molecule must be of sufficient duration to permit the FRET to occur.[7]

The rate of FRET between two chromophores exhibiting dipole-dipole interactions is expressed by the following equation:[8]

𝑘_{𝐹𝑅𝐸𝑇} = 1 τ_𝐷 9000(ln 10)𝑘2ɸ𝐷𝐼 128𝜋5_𝑁𝑛4 1 𝑅6

(23)

3

Figure 1.2. Jablonski diagram of FRET.[7]

1.1 From natural to artificial light-harvesting complexes

There is a wide variety of light-harvesting complexes in nature. They differ in the arrangements of chromophores, chromophore types (e.g. chlorophyll, bilins and carotenoids) and optical properties. Some major examples of photosynthetic antennas are the green sulfur bacterial chlorosome,[9]_{cyanobacterial phycobilisomes,}[10]_{dinoflagellate}

peridinin-chlorophyll protein, [11]_{the Pcb protein (Prochlorococcus chlorophyll a2/b2),}[12]

light-harvesting complex II (LHCII) in green plants[13]_{and the purple bacterial light-harvesting}

complexes 1 (LH1) and 2 (LH2).[14]

Figure 1.3. displays structures of some of the main classes of photosynthetic antenna complexes. All these antennas are membrane proteins containing a network of interacting chromophores. Even though they show structural diversity, all antenna complexes are able to convert the photogenerated excitations to charge separationwith very high efficiency.[15]

Figure 1.2. Some examples of photosynthetic light-harvesting antenna complexes. From left: dinoflagellate peridinin-chlorophyll protein,[11]_{light-harvesting complex II in green plants}[13]_{and the purple bacterial} light-harvesting complexes 1 (LH1) and 2 (LH2).[14]

(24)

4

in bacteria. The unique biological and chemical functions of those molecules are determined by their molecular structures, containing highly delocalized conjugated molecular orbitals. The absorption properties of these chromophores are highly tunable and vary owing to the extent of the conjugation and the number and nature of the substitutions, creating a “rainbow” of colors that covers the visible and the near infrared regions of the solar spectrum (Figure 1.4).[16]

Figure 1.3. Top: General structure of the main pigments used in natural light harvesting chlorophylls (Chl) and bacteriochlorophylls (BChl), which are differently substituted tetrapyrroles. Bottom: Absorption spectra of photosynthetic pigments in various solvents.[17]

Key parameters that characterize antennas are the absorption cross-section of the solar energy, as well as its conversion efficiency and rate. Chromophores must strongly absorb visible or near infrared light (characteristic ε ~ 100,000 M–1cm–1),[3] and the excited states

generated by this absorption must be sufficiently long lived. The efficiency is determined by the fraction of the absorbed energy that reaches the reaction center and can be described by the quantum yield. Commonly, natural antennas present very high quantum yields, not lower than 95%.[4,18]_{Moreover, they are relatively stable supramolecules and are arranged in a}

controlled and precise way that provides paths for excitons to migrate to the reaction centers. The fast energy transfer between single chromophores (characteristic times vary from 100 to 1000 fs)[19]_{should be combined with the relatively slow (tens of picoseconds) electron}

(25)

5

present also ways of deactivating potentially destructive side products such as triplet states and singlet 02•.

The controlled organization of functional chromophores into highly ordered self-assembled arrays mimicking the characteristics of natural photosynthetic systems have polarized a great interest in material science for various applications.[20–22]_{In this respect, supramolecular}

assemblies of spatially well-organized dyes can facilitate excitation energy transfer and are fundamental to the realization of efficient artificial antennas.

First of all, the most important requirement for a light-harvesting system is its capacity to absorb light, the absorption spectrum of its components should cover a substantial part of the visible spectral region. In this respect, the most important step in the design of an artificial antenna system is the selection of the chromophores.

The chromophore is the simplest unit of the light-harvesting complex, but no single chromophore is capable of efficiently powering artificial photosynthesis. For instance, Zn (II) tetraphenyl porpyrin (ZnTPP) has a very high extinction coefficient (500  103_M-1_cm -1_),[23]_{but it covers only a narrow region of the solar spectrum (maximum absorbance peak}

at 424 nm). This problem can be overcome by working with dye aggregates,[24]_{which owing}

to excitonic couplings can have much broader absorption ranges. On the other hands, perylene bisimides (PBIs) have much wider spectral coverage, but present lower molar extinction coefficient than ZnTPP (95  103_M-1_cm-1 _{for unsubstituted perylene bisimide).}[23]

Hence, the incorporation of different chromophores into an organized spatial arrangement that enables efficient energy transfer between them is the second key step for the creation of artificial antenna. For the assembly of the chromophores, covalent and non-covalent strategies have been envisaged allowing to obtain an unlimited number of artificial antennas structurally organized as dendrimers,[25–29]_macrocycles,[30,31]_{supramolecular polymers,}[32]

nanostructures[33–36]_{or assembled using biomaterials such as proteins}[37,38]_{or nucleic}

acids,[39,40]_{having unique photophysical and optical properties.}

An example of covalently linked light-harvesting assembly is presented in Figure 1.4. It is a dendrimer macromolecule in which the peripheral chromophores are the primary light absorbers as in the natural antenna. Dendrimers are well-defined, tree-like macromolecules with a high degree of order and the possibility to contain selected chromophoric units in predetermined sites of their structure and have attracted great attention as novel nanoscopic light-harvesting molecules.[41]_{Müllen and co-workers reported an example of}

(26)

6

multichromophoric rigid polyphenylenic dendrimer bearing a terrylenediimide (TD) as a core, perylenemonoimide (PMI) chromophores in the scaffold and naphtalenemonoimides (NMI) at the rim absorbs light over the whole visible range of the spectrum and emits mainly in the red region due to unidirectional FRET from NMI and PMI donors to TD. [42]_{In fact,}

the spatial positioning of these chromophores within the triad and their respective spectral properties make this multichromophoric system an efficient light collector. It is possible to design and synthesize dendrimers containing a large variety of chromophoric groups organized in the dimensions of time, energy and space in order to obtain efficient light-harvesting devices.[41]

Figure 1.4. Example of a polyphenylene dendrimer with multiple peripheral peryleneimide, naphtalenemonoimides (NMI) at the rim and a central terrylenediimide chromophore.[42]

On the other hands, based on non-covalent chromophore assembly, Kobuke and co-workers described the first example of a cyclic hexamer of phorphyrins prepared by connecting slipped-cofacial dimer units via coordination from the imidazolyl arm to the central Zn ion (Figure 1.5).[30]_{Phorphyrins are ideal chromophores for artificial light harvesting because}

they are structurally similar to photosynthetic pigments and have advantageous photophysical properties, such as photo-stability, visible-light absorption,[43]_long-lived

excited states, rapid excitation energy exchange,[44]_{and high molar extinction coefficients,}

which can be easily tuned by metal coordination.[45]_{In the porphyrin cyclic hexamer}

(27)

7

Figure 1.5. A structural model of the target porphyrin hexameric macroring.

In summary, the choice of the chromophores and the way to assembly them remain the two important steps for the creation of artificial antennas. Even though natural pigments offer us a large panel of photophysical and optical properties, the preparation of new dyes with tunable colors remains a very challenging task for the development of highly efficient light-harvesting materials.

1.2 Tuning the colors of molecules by tuning the HOMO-LUMO gap (HLG)

The design of functional material through band gap and energy level tuning is crucial in developing new dyes with unique photophysical and optical properties. In this respect, various structural variables have to be mastered in order to control the HOMO-LUMO gap of π-conjugated systems. In this section we will briefly introduce the key parameters that have to be taken in consideration to design functional organic chromophores with tunable opto-electronic properties.

The band gap of a material (Eg) derived from a linear π-conjugated system can be defined by the sum of five contributions as described by the following equation, demonstrating that the structural feature of π-conjugates is crucial on the control of the HOMO-LUMO gap.[46,47]

(28)

8

One of the main features that influence the magnitude of the energy gap is explained by the

Peierls instability, which depends on the degree of Bond length alternation (BLA) in the conjugated path. Specifically, EBLA is related to the difference between single and double bond lengths. Synthetic modifications leading to structural changes resulting in a reduced BLA can be expected to decrease the HOMO-LUMO gap (HLG).[46] _{For instance, in}

aromatic systems like poly(p-phenylene), poly(pyrrole) or polythiophene (Figure 1.6), that have a non-degenerated ground state, the two limiting mesomeric forms (aromatic vs quinoid) obtained by the flip of the double bonds are not energetically equivalent. In most cases the quinoid form is characterized by a smaller Eg.[48]

Figure 1.6. Schematic representation of the two limiting mesomeric forms obtained by the flip of double bonds for poly(p-phenylene), poly(pyrrole) or polythiophene .[46]

(29)

9

intermolecular interactions can affect the magnitude of the band gap. The described structural factors that determine the band gap of linear π-conjugated systems are summarized in Figure 1.7.

Figure 1.7. Representation of the structural factors that influence the band gap of materials derived from π-conjugated systems.[47]

In the present work we will discuss in detail the use of different synthetic tools related to one or several structural contributions discussed above to finely tune the HOMO-LUMO gap of π-conjugates. In particular, an approach to control the energy band gap that involves the elongation of the aromatic surface along with planarization and rigidification of the conjugated system will be employed.

1.3 Outline of the dissertation

(30)

10

Figure 1.8. Schematic representation of the outline of this doctoral dissertation

Chapter II describes the design and the synthesis of a novel oligo (p-phenylenevinylenes)

chromophore for the functionalization of single walled carbon nanotubes (SWCNTs). Since photoinduced processes attracted much interest because of their relevance to solar energy conversion systems, we will present the preparation and the photophysical characterizations of new systems made of accepting single chirality enriched SWCNTs and electron-donating dimethylamino distyrylbenzene derivatives. A covalent approach via amidation reaction has been exploited to anchor the chromophoric unit to the SWCNT.

(31)

11

1.4 Bibliography

[1] N. Armaroli, V. Balzani, Angew. Chemie - Int. Ed. 2007, 46, 52–66. [2] N. S. Lewis, MRS Bull. 2007, 32, 808–820.

[3] G. D. Scholes, G. R. Fleming, A. Olaya-Castro, R. van Grondelle, Nat. Chem. 2011,

3, 763–774.

[4] B. Green R., W. Parson, Advances in Photosynthesis and Respiration -

Light-Harversitng Antennas in Photosynthesis., 2003.

[5] S.-S. Sun, N. S. Sariciftci, Organic Photovoltaics, Taylor & Francis, 2005. [6] D. Gust, T. A. Moore, A. L. Moore, Acc. Chem. Res. 1993, 26, 198–205. [7] S. A. Hussain, Energy 2009, 132, 1–4.

[8] G. D. Scholes, Annu. Rev. Phys. Chem. 2003, 54, 57–87.

[9] J. Pšenčík, T. P. Ikonen, P. Laurinmäki, M. C. Merckel, S. J. Butcher, R. E. Serimaa, R. Tuma, Biophys. J. 2004, 87, 1165–1172.

[10] N. Adir, Photosynth. Res. 2005, 85, 15–32. [11] T. Larkum, Trends Plant Sci. 1996, 1, 247–248.

[12] C. S. Ting, G. Rocap, J. King, S. W. Chisholm, Trends Microbiol. 2002, 10, 134–142. [13] Z. Liu, H. Yan, K. Wang, T. Kuang, J. Zhang, L. Gui, X. An, W. Chang, Nature 2004,

428, 287–292.

[14] R. Van Grondelle, V. Novoderezhkin, Biochemistry 2001, 40, 15057–15068.

[15] R. J. Cogdell, A. T. Gardiner, H. Hashimoto, T. H. P. Brotosudarmo, Photochem.

Photobiol. Sci. 2008, 7, 1150–8.

[16] T. Szabó, M. Magyar, K. Hajdu, M. Dorogi, E. Nyerki, T. Tóth, M. Lingvay, G. Garab, K. Hernádi, L. Nagy, Nanoscale Res. Lett. 2015, 10, 458.

(32)

12

[20] G. M. Whitesides, J. P. Mathias, T. Seto, Christoph, Science. 1991, 254, 1312. [21] D. Philp, J. F. Stoddart, Angew. Chem. Ed. Engl. 1996, 35, 1154–1196.

[22] L. Maggini, D. Bonifazi, Chem. Soc. Rev. 2012, 41, 211.

[23] P. D. Frischmann, K. Mahata, F. Würthner, Chem. Soc. Rev. 2013, 42, 1847–70. [24] Z. Chen, A. Lohr, C. R. Saha-Möller, F. Würthner, Chem. Soc. Rev. 2009, 38, 564–

584.

[25] M. S. Choi, T. Yamazaki, I. Yamazaki, T. Aida, Angew. Chem. Int. Ed. 2004, 43, 150–158.

[26] M.-S. Choi, T. Yamazaki, I. Yamazaki, T. Aida, Angew. Chem. 2004, 116, 152–160. [27] S. Hecht, J. M. J. Frechet, Angew. Chem. Int. Ed. 2001, 40, 74–91.

[28] U. Hahn, M. Gorka, F. Vögtle, V. Vicinelli, P. Ceroni, M. Maestri, V. Balzani, Angew.

Chem. Int. Ed. 2002, 41, 3595–3598.

[29] V. Balzani, G. Bergamini, P. Ceroni, E. Marchi, New J. Chem. 2011, 35, 1944–1954. [30] R. Takahashi, Y. Kobuke, J. Am. Chem. Soc. 2003, 125, 2372–2373.

[31] P. Parkinson, C. E. I. Knappke, N. Kamonsutthipaijit, K. Sirithip, J. D. Matichak, H. L. Anderson, L. M. Herz, J. Am. Chem. Soc. 2014, 136, 8217–8220.

[32] R. Abbel, C. Grenier, M. J. Pouderoijen, J. W. Stouwdam, P. E. L. G. Lecle, R. P. Sijbesma, E. W. Meijer, A. P. H. J. Schenning, J. Am. Chem. Soc 2009, 131, 833– 843.

[33] T. E. Kaiser, H. Wang, V. Stepanenko, F. Würthner, Angew. Chem. Int. Ed. 2007, 46, 5541–5544.

[34] A. Lohr, F. Würthner, Angew. Chem. Int. Ed. 2008, 47, 1232–1236.

[35] A. Ajayaghosh, V. K. Praveen, C. Vijayakumar, S. J. George, Angew. Chem. Int. Ed.

2007, 46, 6260–6265.

[36] A. Ajayaghosh, C. Vijayakumar, V. K. Praveen, S. S. Babu, R. Varghese, J. Am.

Chem. Soc. 2006, 128, 7174–7175.

(33)

13

[38] M. Endo, M. Fujitsuka, T. Majima, Chem. A Eur. J. 2007, 13, 8660–8666.

[39] P. K. Dutta, R. Varghese, J. Nangreave, S. Lin, H. Yan, Y. Liu, J. Am. Chem. Soc.

2011, 133, 11985–11993.

[40] J. G. Woller, J. K. Hannestad, B. Albinsson, J. Am. Chem. Soc. 2013, 135, 2759– 2768.

[41] V. Balzani, P. Ceroni, M. Maestri, V. Vicinelli, Curr. Opin. Chem. Biol. 2003, 7, 657– 665.

[42] M. Cotlet, T. Vosch, S. Habuchi, T. Weil, K. Müllen, J. Hofkens, F. De Schryver, J.

Am. Chem. Soc. 2005, 127, 9760–9768.

[43] S. Prathapan, T. E. Johnson, J. S. Lindsey, J. Am. Chem. Soc. 1993, 115, 7519–7520. [44] Y. Terazono, G. Kodis, K. Bhushan, J. Zaks, C. Madden, A. L. Moore, T. A. Moore,

G. R. Fleming, D. Gust, J. Am. Chem. Soc. 2011, 133, 2916–2922.

[45] K. Tomizaki, R. S. Loewe, C. Kirimaier, J. Schwartz, J. Retsek, F. Bocian D, D. Holten, S. J. Lindsey, J. Org. Chem. 2002, 67, 6519–6534.

[46] J. Roncali, Chem. Rev. 1997, 97, 173–206.

(34)

14

2. Oligo(p-phenylene vinylene) chromophores: from the synthesis to the

functionalization of SWCNTs for photovoltaic applications

This chapter describes the preparation and physical characterization of oligo(p-phenylene vinylene)-derived single-walled carbon nanotubes (SWCNTs). The first part addresses the synthesis and characterization of oligo(p-phenylene vinylenes) (OPV) modules and the covalent functionalization of SWCNTs enriched in (7, 6) chirality. The second part sets out the photophysical properties of OPV functionalized carbon nanotubes and the understand of photo-induced interactions between SWCNTs and selected molecular system.

The chapter is divided in three main sections: i) section 2.1 includes a general introduction on carbon nanotubes giving an overview on their chemistry and properties, along with a brief overview about organic photovoltaics, ii) section 2.2 deals with the design of OPV-based donor-acceptor systems for photovoltaic devices and describes the synthesis of the targeted OPV-functionalized carbon nanotubes; iii) section 2.3 describes the absorption and emission properties of OPV decorated SWCNT, while iv) section 2.4 describes the characterization of OPV-SWNT as doping agents for spiro-OMeTAD in a flat CH3NH3PbI3-based solar cell.

The research work in section 2.3 has been carried out jointly with Eleonora Pavoni from the group of Prof. Nicola Armaroli at the Istituto per la Sintesi Organica e la Fotoreattività (CNR - ISOF), Bologna, Italy. The research described in sections 3.4 was performed by

Vanira Trifiletti,Aurora Rizzo,Dr.Andrea Listorti and Dr. Silvia Colella from the group of Dr. Gianluca Accorsi at the Istituto di Nanotecnologie (CNR-Nanotec) and Università del Salento, Lecce, Italy.

The X-ray analysis presented in this chapter was performed by Nicola Demitri (Elettra –

(35)

15

2.1 General introduction on Oligo(p-phenylene vinylene) based Donor-Acceptor Systems for Photovoltaic Devices

The unique, nanosized, three-dimensional structure of natural photosynthetic systems and the functions of solar-energy conversion have incited the scientific community to mimic such processes in artificially organized systems. Since, the reaction centers of photosynthetic organisms can be assumed as photovoltaic devices at the molecular level, donor and acceptor molecules have been organized onto electrode surfaces to obtain details on the relationship between the surface structure and the photo-electrochemical properties.[1]_{In this respect,}

recent studies have used carbon nanotubes (CNTs) as nanoplatforms to attach molecules, providing hybrid systems that can be designed for specific applications, including light-harvesting devices.[2]_{CNTs present outstanding properties such as high conductivity, strong}

mechanical, thermal and environmental stability.[3]_{Despite these unique properties of CNTs,}

one of the major drawbacks is that they are often produced with metal catalyst and amorphous carbon impurities. Since these impurities can obscure the properties of CNTs, they need to be removed, nitric acid treatment and Air/HCl treatment being the most commonly used purification procedures.[4,5]_{These treatments have been shown to create}

oxygen-containing functional groups on the CNT surfaces that can be used as reaction sites for a further chemical derivatization.[6,7]_{The covalent attachment of molecules to CNTs}

broadens the range of potential applications of CNT systems and allows the tailoring of the properties of modified CNTs. In the recent years a lot of attention has been addressed to the preparation of multicomponent systems and nanohybrids exploiting the covalent attachment of electron donors such as ferrocene,[8]_{phthalocyanines}[9]_{or porphyrins}[10,11]_onto

nanocarbon materials. For example, adducts of single walled carbon nanotubes (SWCNTs) with pyrene or porphyrin exhibit fast electron transfer, leading to long-lived charge separated states.[12]

In this chapter we reported the design and preparation of SWCNTs based donor-acceptor system for the construction of molecular photovoltaic devices. The acceptor moiety is represented by SWCNTs enriched in 7, 6 chirality, while for the donor moiety a π-conjugated oligo(p-phenylene vinylene) has been chosen. The choice of the latter is motivated by the stability, high luminescent efficiency, and ease of synthesis of those compounds.[13,14]_The

(36)

16

2.1.1 General Introduction on Organic Photovoltaic Cells

The ever-increasing quest for sustainable and renewable sources of energy promoted enormous interest in the solar cell technology. Solar energy is the most abundant, inexhaustible and clean source of energy till date. The power from the sun intercepted by the earth is many times larger than the present rate of the world energy consumption.[15]_{In this}

respect, solar photovoltaic technology represents one of the finest ways to harness the solar power. Photovoltaic devices deal with the conversion of sunlight into electrical energy. They belong to three main classes: organic (e.g., heterojunction cells), photoelectrochemical cells (e.g., dye sensitized solar cells) and inorganic (e.g. silicon), the latter being used in the majority of the solar cells in the market today. Such solar cells have been highly optimized and show efficiencies greater than 15 %.[16]_{Nevertheless, their high manufacturing costs,}

difficulties in tailoring the cells to various applications and the inherent physical rigidity of silicon hamper their widespread use as solar-based power. On the other hand, solar cells, such as dye-sensitized solar cells (DSSCs)[17,18]_{and bulk heterojunction cells (BHJs)}[19–21]

are promising for inexpensive and large-scale solar energy conversion.

The study of such devices started by the discovery of the photovoltaic effect by the French Scientist E. Bequerel in 1839[22]_{who observed an electric potential between two electrodes}

attached to a solid or liquid system upon light irradiation. This breakthrough has been the base for a variety of concept to convert solar radiation into electricity, paving the way for new alternative energy generation.

The main processes of converting light into electric current in all organic photovoltaic cells are (i) absorption of a photons leading to the formation of an excited state, the electron-hole pair (exciton generation), (ii) Exciton diffusion to a region, where (iii) charge separation occurs. Finally (iv) the charge transport to the anode (holes) and cathode (electrons) to supply a direct current for the consumer load.[23]_{The organic photovoltaic cells can be}

constructed in a variety of ways, including single layer, bilayer heterojunction and bulk heterojunction cells[24]_{(Figure 2.1). In spite of the different device architectures, the}

principles of their operation are substantially the same.[25]

(37)

17

completely different materials. Each material has a characteristic HOMO and LUMO energy, with the HOMO and LUMO of the donor being higher in energy than their respective counterparts in the acceptor. The gap between the HOMO and the LUMO is referred as the optical band-gap and determines the minimum wavelength of light required for excitation in each material.

The schematic energy-band diagram of a donor–acceptor heterojunction is reported in Figure 2.1 b. In order to allow the interaction of light with the photovoltaic material, the device is constructed onto a transparent substrate (glass, plastic film) and one of the conducting contacts is also transparent, so the material can absorb the incident sunlight photons and generate excitons. An exciton is an excited state corresponding to a bound state of an electron, e−_{, and an imaginary particle called an electron hole, h}+_{. Upon irradiation, an}

electron excitation from the HOMO to the LUMO of the donor generates an exciton, leaving behind a hole of opposite charge (process 1 in Figure 2.1 b). Since the promoted electron and the hole have opposite electric charges, both remain bounded by a Coulomb force, making this state slightly more stable than a dissociated free electron and a hole. It is noteworthy to indicate that only photons higher in energy than the band gap (hν ≥ Eg) lead

to the photogeneration and that the energy transferred to the exciton increases with Eg (i.e.,

higher is the energy content of the photon, higher is the energy of the exciton generated). The exciton is then dissociated into charges of different sign (process 2 in Figure 2.1 b) generating an electron lying in the LUMO of the acceptor material and a hole in the HOMO of the donor material. [23]_{The overall electric field between the two electrodes drives}

(38)

18

Figure 2.1 (a) Typical heterojunction organic solar cell architecture; (b) schematic energy-band diagram of a donor–acceptor heterojunction. (c) schematic bulk-heterojunction solar cell. Donor and acceptor materials are intimately mixed together to form a dispersed heterojunction.[23]

While in homojunction devices exciton dissociation usually takes place at the junction with the electrodes, in heterojunction and bulk-heterojunction devices it takes place much more efficiently at the donor–acceptor interface, leading to a free electron in the acceptor material and a free hole in the donor. Since donor materials are more likely to conduct holes and acceptor materials electrons, the bulk-heterojunction architecture ensures the transport of charge carriers to the electrodes with only a small chance to recombine with their counterpart (e.g. bumping into each other, process 5 in Figure 2.1 b) as they do not have to diffuse through the same material (in contrary to homojunction devices), and connect with the correct electrode. To increase the close contact between the donor and the acceptor materials and to provide complete transport of excitons to the donor-acceptor interface and thus exciton dissociation, the bulk-heterojunction architecture was developed.[23]

The first significant reports regarding organic photovoltaic cells date back to beginning of the 20th_{century. In 1959, Kallaman and Pope}[26]_{reported on photovoltaic measurements on}

thin (10 µm) anthracene single crystals cell. The anthracene crystal was positioned between two NaCl solutions, which acted as transparent electrical contacts and were further connected to silver electrodes. They reported an efficiency of only 2 x 10-6_{%. Later, the}

same authors also observed a photovoltaic effect in a tetracene water system.[27]_Eventually,

in 1978 Fishman and co-workers reported on organic solar cells based on merocyanine dyes that exhibited sunlight efficiencies in excess of 1 %.[28]_{However, these single layer cells}

(39)

19

polymer/polymer,[32,33]_polymer/dye[34–36]_{or dye/dye}[37,38]_{donor-acceptor blends. One of the}

most promising examples concerns the fabrication of cells composed of thieno[3,4-b] thienyl- and benzodithienyl-based polymers (PTBs) as donor specie and fullerene phenyl-C71-butyric acid methyl ester (PC71BM) as acceptor, yielding above 7.4 % power conversion

efficiency under ambient conditions.[39]

The design and the choice of the donor and acceptor molecules represented one of the most important steps in the construction of photovoltaic devices. The selected molecules must be characterized by optical band-gaps that are small enough to be excited efficiently by the solar radiation. Moreover, the donor molecule has to stabilize the hole transport, while the acceptor must favor the electron transport. The combination of the two must also be chosen to allow for the proper offset between their HOMO and LUMOs. Finally, the molecules must be properly functionalized in order to be solubilized for processing.

Based on all that, the development in the mid-1990 of solar cells based mainly on photoactive organic materials, has offered the prospect to make solar power affordable for far broader uses. Donor-acceptor based organic solar cells are currently showing power conversion of more than9 %.[40]_{Nevertheless, efficiencies of these organic devices have not yet reached}

those of their inorganic counterparts (10-25 %).[40]_{Despite this, organic photovoltaic}

technology presents some great advantages: i) the use of low quantities of materials; and ii) the possibility of tuning the optoelectronic properties of the organic molecules involved by chemical modification; iii) the composition based on handy materials and the fact that the fabrication process does not need high temperatures, allowing the generation of very flexible thin films with plastic substrates.

In order to reach higher efficiencies, innovative materials such as carbon nanotubes (CNTs) have been integrated in organic photovoltaic cells.[41]_{CNTs are one-dimensional}

nanostructures characterized by a ballisitic charge transport along their axis. In addition, their band-gap can be tuned by employing different radii and chiralities, allowing a precise band engineering. In addition, thanks to their high surface area (about 1600 m2_g-1_{) and their}

electron-accepting properties, CNTs can be considered as a wide conductive network offering a tremendous opportunity for exciton dissociation. For these reasons, CNTs have been widely studied in electro- and photo-active nanocomposites in association with semi-conducting materials.[42,43]

(40)

poly(3-20

octylthiophene) (P3OT),[43] or poly(p-phenylenevinylenes) (PPV),[44] into ITO based

photovoltaic devices. However, CNTs present different drawbacks that limit their effectiveness in multifunctional CNTs-based nanoconjugates and/or nano-hybrid photovoltaic systems. Key issues are addressed to the lack of a reliable and reproducible control of the surface chemistry and their organization within the active layer due to their particular intermolecular cohesive forces (0.5 eV nm-1_{). In fact, CNTs tend to strongly}

aggregate both in the solid state and in solution, forming bundled structures that impair the reproducibility of the conducting properties and display a reduced threshold of a few percent. An efficient strategy to bypass this and create easy-to-process CNT-based materials employs their chemical derivatization,[45]_{either in a covalent}[46–49]_{or non-covalent}[50–53]

fashion. Specifically, the incorporation of photoactive antenna chromophores (displaying high extinction coefficient in the visible region of the solar spectrum) such as porphyrins[1]

emerged as one of the most suitable routes to engineer charge-separation and photovoltaic conversion.

2.1.2 General Account on Carbon Nanotubes

Carbon structures can exhibit multiple allotropic forms, with very diverse properties from the soft and conductive graphite to the hard and insulating diamond. Fullerenes, are molecules composed entirely of carbon arranged in hexagonal and pentagonal rings assuming the form of a hollow sphere. Carbon nanotubes (CNTs) are another allotropic form of carbon and, when synthesized, are often capped by a half fullerene at the end.[54]_The

discoveries of C60 in 1985 by Kroto et al.[55] and carbon nanotubes in 1991 by Ijima and

co-workers[56]_{have inspired a new interdisciplinary era in material science and technology.}

Both fullerene and carbon nanotubes (CNTs)[57]_{display unique structures that bring with}

them remarkable mechanical, thermal, and optical properties that are extremely promising for applications in electronics, advanced materials and medicinal chemistry.[58,59]

2.1.2.1 CNTs structure and electronic/optical properties

(41)

21

Figure 2.2 Representation of single walled carbon nanotube (SWCNTs). It can be pictured as a rolled graphene sheet and according to the rolling direction can be classified as zigzag, armchair or chiral.[45]

There are two main types of carbon nanotubes presenting high structural perfection: Single-Walled Carbon Nanotubes (SWCNTs) that consists of a single graphite sheet seamlessly wrapped into a cylindrical tube (Figure 2.2) and Multi-Walled Carbon nanotubes (MWCNTs) that comprise an array of single walled nanotubes concentrically nested reaching diameters of up to 100 nm.[6]

Most SWCNTs present a diameter about 0.4 to 4 nm, with a tube length that can be many millions of times longer, and due to the high ratio between the two dimensions they are virtually considered mono-dimensional (1D) objects. Despite structural similarity to a single sheet of graphite, which is a semiconductor with zero band gap, SWCNTs may be either metallic or semiconducting, depending on the rolling direction of the graphene sheet (Figure 2.3).[60]_{Thus, nanotubes have different structures, which can be described by the chiral}

vector (n, m), where the integers n and m represent the number of unit vectors along two directions in the honeycomb crystal lattice of graphene and are described by the equation: 𝐶ℎ = 𝑛𝑎1+ 𝑚𝑎2. The chiral vector is determined by the diagram in Figure 2.3. The (n,m)

indices fullydefine the SWCNT radius and chirality and determine its electronic structure.

(42)

22

Figure 2.3. A 2D graphene sheet showing chiral vector Ch and chiral angle θ,[58]_{(left) that give rise to armchair,} zig–zag and chiral nanotube structures with metallic or semi-conducting electronic character (right).[61] All armchair SWCNTs are metallic (|n-m| = 3q), while those with |n-m| ≠ 3q, where q is a

nonzero integer, are semiconductors with a tiny band gap. All the others are semiconductors with a band gap that inversely depends on the nanotube diameter.[62]_{Usually, without}

chirality control, we obtain one third metallic and two thirds semiconducting SWCNTs. Eventually, we can underline that the chiral vector, (n, m), determines both the diameter of a SWCNT and its opto-electronic properties. In fact, the three SWCNTs depicted in Figure 2.3 have approximately the same tube diameter, but they dramatically differ for their opto-electronic properties. The opto-electronic density of states (DOS) of a SWCNT (Figure 2.4) consists of distinctive levels known as van Hove singularities (vHS) defined by circumferential wave vectors. As already mentioned, all armchair SWCNTs are metallic with a continuous DOS near the Fermi level (EF) (which is highlighted by the dashed blue line in Figure 2.4), while the DOS of the semiconducting SWCNTs show a significant band gap on the order of 500 meV that varies inversely with diameter. Metallic SWCNTs are characterized by higher dielectric constants (ε >1000) than the semiconducting one (ε < 10).[63]_{On the other hand, larger diameter SWCNTs present larger dielectric constants than}

smaller ones. The position of the Fermi level, as well as the reduction and oxidation potentials, also vary as a function of diameter.

(43)

23

Figure 2.4 Schematic view of the electronic density of states (DOS) of metallic and semiconducting SWCNTs.[61]

A SWCNT is considered perfectly crystalline i.e. without defects if the graphene sheet has no alteration in the hexagonal aromatic structure of the carbon atoms along the tube. However, CNTs are not absolutely perfect due to the presence of certain defects. They can be mainly classified into four groups: i) topological defects, corresponding to the presence of rings other than hexagons, for example pentagon/heptagon pairs[64]_{ii) rehybridization}

(chemical treating defects consisting of atoms/groups covalently attached to the carbon lattice of the tube); iii) incomplete bonding defects (vacancies, dislocations, etc); iv) doping with other elements than carbon. Either they can be strategically inserted, for instance treating CNTs with very harsh acidic conditions,[65]_{either they can be inevitably formed}

during the growth of CNTs. In any way, their presence in the carbon network can lead to attractive properties and new potential nano-devices. Defects are also key factor in the covalent chemistry of CNTs because they can serve as anchor groups for further functionalization.

Carbon nanotubes can be produced using several different techniques while new routes are continuously being developed. Commonly used methods exploit the catalytic decomposition of certain hydrocarbons as source of carbon on small metal particles (e.i. Fe, Co, Ni) under a certain applied energy. The diameter of so prepared nanotubes is governed by that of the catalyst particles responsible for their growth. The most common methods are: arc discharge of graphite, laser ablation of carbon and chemical vapor deposition techniques,[54]_that

(44)

24

2.1.2.2 Chemistry of Carbon Nanotubes

As discussed before, SWCNTs tend to form bundles due to the presence of strong van der Walls interactions (0.5 eV nm-1_{) between tube-to-tube contacts. This feature results in scarce}

solubility in almost all organic solvents and aqueous solutions, which is an obstacle for the full exploitation of their properties. For these reasons, to be used into any type of devices, an appropriate modification has to be undertaken in order to render them dispersible and thus processable. Several protocols have been described in the literature to exfoliate and functionalize SWCNTs in order to obtain individuals and/or small bundles of carbon nanotubes.[7]_{They can be chemically modified mainly by the following strategies: (i)}

functionalization of the defect located on the sidewall and the rims; (ii) non-covalent interactions; (iii) sidewall covalent functionalization and (iv) endohedral inclusion.[66]

The covalent functionalization of nanotubes is more robust and better controllable compared to functionalization based on non-covalent method, and therefore will be adopted in this research work. Carbon nanotubes structure can be divided into two main regions: the end caps and the side wall. The end caps of a carbon nanotube resemble a hemispherical fullerene and its reactivity. In fact, the reactivity of the fullerenes is primarily driven by the enormous strain generated by their spherical geometry as reflected in the pyramidalization angles (θp) of the carbon atoms (Figure 2.5 a).[67]_{For an sp}2_{-hybridized (trigonal) carbon atom, planarity} is strongly preferred, and this implies a pyramidalization angle of 0°, while in contrast an sp3_{-hybridized (tetrahedral) carbon atom requires an angle of 19.5°. In the case of fullerene}

all carbon atoms have an angle of 11.6°, indicating that their geometry is more appropriate for tetrahedral than trigonal hybridization. Thus, the chemical conversion of any trivalent carbon atom in C60 to tetravalent carbon is accelerated by the strain relief that strongly favors

the addition chemistry on fullerene molecules.[67]_{The side wall, instead, presents carbon}

atoms with a lower pyramidalization angle, but with the p-orbital misalignment as major source of strain in that region. In fact, a carbon nanotube with the same radius as C60 exhibits

(45)

25

Figure 2.5. (a) pyramidalization angle (θP), and (b) the π-orbital misalignment angles along the C1-C4 in the (5,5) SWNT and its capping fullerene, C60.[67]

The reactivity of carbon nanotubes can be rationalized in terms of curvature-induced pyramidalization angle and misalignment (ϕ) of the π-orbitals in comparison with the graphene structure. Both factors induce a local strain, leading to CNTs generally more reactive than a flat graphene sheet. Furthermore, on the bases of the carbon nanotubes structure there is a correlation between tube diameter and reactivity. Both pyramidalization angles and p-orbital misalignment are inversely proportional to the tube diameter, then smaller carbon nanotubes are expected to be more reactive than larger nanotubes. Moreover, the CNTs reactivity depends also on the chirality of the system as reported by Li and co-workers[68]_{and Kataura and co-workers.}[69,70]_{Considering SWCNTs with the same}

diameter: the zig-zag carbon nanotubes are less reactive than armchairs that are less reactive than the chiral tubes.

Two main approaches have been developed for the covalent functionalization of carbon nanotubes: (i) the addition chemistry to SWCNTs; and (ii) amidation and esterification of oxidized SWCNTs.[7]_{The direct functionalization of pristine SWCNTs can be obtained via}

electrophilic, nucleophilic or radical additions using highly reactive species. Several reviews have been reported on the chemistry of SWCNTs describing exhaustive or representative examples of addition reactions.[59,71,72]_{On the other hand, the treatment of CNTs under}

(46)

26

and sulfuric acid, or heating in a mixture of sulfuric acid and hydrogen peroxide, results in the formation of short and opened tubes bearing oxygenated functions such as carbonyl, carboxyl, hydroxyl groups. These functional groups can then serve as anchor point for further functionalization. For instance, the acid functions can easily react with alcohols or amines to give ester or amide derivatives.[7]

Amidation or esterification reactions can be carried out on oxidized SWCNTs by standard methods, either using acid chlorides as intermediates or carbodiimide-based coupling reagents. This approach offers the possibility of attaching many organic fragments and synthesizing a great variety of CNT derivatives. Haddon and coworkers[73]_{first reported}

different functionalization strategies leading to soluble SWCNTs via amidation reactions between oxidized nanotubes and octadecylamine (Figure 2.6, 2-12) and 4-tetradecylaniline through the formation of an acyl chloride intermediate. The octadecylamine groups present on the nanotubes increased the solubility of the material in most of organic solvents, assisting its characterization and facilitating the achievement of highly purified SWCNT that are suitable for physical properties measurements.

This approach has been widely exploited to afford a large variety of functionalized CNTs and the most significant results have been collected in a variety of outstanding review papers.[7,66,71,72]_{Herein we report some examples where this approach has been used for}

combining the SWCNT properties with those of other interesting materials. For instance, coupling SWCNTs with C60 fullerene create unique structures that have been studied for

charge-transfer properties. In 2007, Langa and co-workers described the first synthesis of hybrid conjugated SWCNT–C60 materials 2-13 (Figure 2.6). The authors decorated the

SWCNTs with fragments of N-anilinopyrazolino C60 fullerene, by an amidation reaction

with the SWCNT acyl chloride. Similarly, a grapevine nanostructure (2-14) was prepared, based on SWCNT covalently functionalized with C60 via amidation reaction.[74] This hybrid

material was investigated by means of Electron Spin Resonance (ESR) spectroscopy revealing the presence of an electron transfer process between the SWCNT and the fullerene. Using the same synthetic methodology, Bonifazi et al. reported the synthesis of several hybrid fullerene derivatives–SWCNT materials that combined C60 fullerenes with appended

photoactive ferrocene or porphyrin functionalities and SWCNTs (2-15, Figure 2.6). X-Ray Photoelectron Spectroscopy (XPS) analysis has been used to prove the presence of C60

(47)

27

appearance of the characteristic photoelectron N 1s emission peak at 400.3 eV, typical of amide groups.

Figure 2.6 Derivatization reactions of acid-cut nanotubes via amidation reaction.

Martin, Prato, Guldi et al.[75]_{also described the functionalization of SWCNTs with the}

strong electron donor tetrathiafulvalene (TTF) (2-16, Figure 2.6) or its π-extended analogs (exTTF) (2-17), through esterification or amidation reactions. This work reports the preparation of the first TTF–SWCNT donor–acceptor systems, to evaluate the possible use of CNTs in solar energy conversion systems. The functionalization was confirmed by different analytical, spectroscopic and microscopic techniques. Photophysical analysis by time resolved spectroscopy revealed the presence of radical species indicating the existence of an efficient photoinduced electron transfer, a critical point for photovoltaic devices. The covalent functionalization of SWCNTs via amidation reaction will be exploited for the prupose of this research work and the decisive synthetic strategy adopted for the functionalization of SWCNTs with small chromophoric molecules is reported in the following section 2.2.

2.1.3 Aim of the project